1. Field of the Invention
The present invention relates to dispersion compensating (DC) optical fibers, and more particularly to single mode DC fibers that are particularly well suited for compensating for dispersion in transmission fibers having kappa values of less than 100.
2. Technical Background
To meet the ongoing drive for more bandwidth at lower costs, telecommunications system designers are turning to high channel count dense wavelength division multiplexing architectures, longer reach systems and higher transmission bit rates. This evolution makes chromatic dispersion management critical to systems performance as system designers now desire the ability to accurately compensate dispersion across entire channel plans. More specifically, the increased demand for higher bit transmission rates has resulted in a large demand for optical transmission systems that can control dispersion effects. An analysis of common optical transmission systems indicates that while such systems can tolerate about 1,000 ps/nm residual dispersion at 10 Gbit/second, these systems tolerate only about 62 ps/nm residual dispersion at a higher transmission rate of 40 Gbit/second. Therefore, it is important to accurately control the dispersion for such high bit rate transmissions. This control becomes increasingly important as the transmission rate increases. Further, the need to accurately control dispersion means that dispersion slope of a transmission fiber must also be compensated for, particularly as transmission rates approach 40 Gbit/second.
Various solutions have been proposed to achieve the low dispersion and dispersion slope values required for compensating non-zero dispersion shifted fibers (NZDSFs), including: photonic crystal fibers, higher order mode dispersion compensation fibers, dispersion compensating gratings and dual fiber dispersion compensating techniques. Each of these solutions have drawbacks associated therewith.
Photonic crystal fibers are designed to have a large negative dispersion and a negative dispersion slope that are close to those required for compensating NZDSFs. However, photonic crystal fibers have drawbacks including a relatively small affective area of about 10 xcexcm2 or less that leads to unacceptably high splice losses and hence requires the use of a specially designed transition or bridge fiber to reduce splice losses. In addition, due to the very nature of photonic crystal fibers, i.e., glass/air interfaces in the core of the fiber, the related attenuation is unacceptable in the transmission window of interest. Further, photonic crystal fibers are significantly difficult to manufacture on a large scale and are, therefore, expensive.
Higher order mode (HOM) dispersion compensation relies on the dispersion properties of a HOM fiber propagating higher order modes. It has been demonstrated that higher order modes, e.g. LP02 and LP11, have higher negative dispersions and dispersions slopes than fundamental modes. HOM dispersion compensation typically relies on the conversion of a transmitted fundamental mode to one of the higher order modes via a mode converter. Subsequently, this higher order mode is propagated in the HOM fiber that supports that mode. After a finite distance, the higher order mode is converted back to the fundamental mode via a second mode converting device. Problems associated with HOM dispersion compensation solutions include inefficient mode converters and difficulty of producing HOM fibers that allow higher order mode transmission while resisting coupling to the fundamental mode.
Dispersion compensating gratings are utilized to achieve a required differential group delay via chirped gratings. Techniques utilizing dispersions compensating gratings have been shown to be useful only for narrow bands, as these techniques typically suffer from dispersion and dispersion slope ripple when the required grating length becomes large.
In dual fiber dispersion compensating fiber solutions for NZDSFs, the dispersion and slope compensation are partially de-coupled and separately treated in each fiber. Typically, dual fiber dispersion compensating techniques include the use of a dispersion compensating fiber followed by a dispersion slope compensating fiber. Such solutions require the use of a dispersion slope compensating fiber that compensates for a relative small dispersion slope. Further, such solutions are subject to additional splice losses and are a comparatively complex solution.
Extensive profile modeling of optical fibers has resulted in well-established correlations between dispersion slope, effective area and bend sensitivity. By increasing the role played by wavelength dispersion in a given fiber, it is possible to decrease the slope and even create a negative slope in some cases. However, as the affective area is decreased, the bend sensitivity of the fiber is increased. The effective area of the fiber can be further increased, but generally at the expense of further degradation of the bend sensitivity. Decreasing the dispersion slope, or making the dispersions slope negative, results in working close to the cut-off wavelength of the fundamental mode, which in turn makes the fiber more bend sensitive and results in greater signal loss at wavelengths greater than 1560 nm. As a result of these relationships, it is extremely difficult to manufacture a viable DC fiber that compensates both dispersion and dispersion slope effectively. Moreover, when kappa (dispersion/dispersion slope) is low, it is difficult to adequately compensate for dispersion over wide wavelength bands.
Heretofore, the most viable broad band commercial technology available to reduce or eliminate dispersion has been DC fiber modules. As dense wavelength division multiplexing deployments increase to 16, 32, 40 and more channels, broad band DC products are desired. Further, as interest increases in higher bit-rate information transmission, i.e., greater than 40 Gbit/second, ultra-long reach systems, i.e., systems greater than 100 km in length, and optical networking, it has been imperative to use DC fibers even on networks that carry data on NZDSFs. One such NZDSF is LEAF(copyright) optical fiber, manufactured by and available from Corning Incorporated of Corning, N.Y. LEAF(copyright) fiber is a positive NZDSF that has been the optical fiber of choice for many new systems due to its inherently low dispersion and economic advantage over conventional signal mode fibers.
The combination of the early versions of DC fibers with NZDSFs effectively compensated dispersion at only one wavelength with significant residual system dispersion being present within the operating band at wavelengths away from that one wavelength. However, higher bit-rates, longer reaches and wider bandwidths require dispersion slope to be compensated for more exactly. Consequently, it is desirable for the DC fiber to have dispersion characteristics such that its dispersion and dispersion slope are matched to that of the transmission fiber it is required to compensate.
Therefore, it would be desirable to develop alternative DC fibers, particularly fibers having the ability to compensate for dispersion and dispersion slope of NZDSFs and other positive dispersion optical fibers over a wide wavelength band of around 1550 nm.